DIET: Learning to Distill Dataset Continually for Recommender Systems
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912982970138624 |
|---|---|
| author | Zhang, Jiaqing Wang, Hao Yin, Mingjia Chen, Bo Jia, Qinglin Zhou, Rui Tang, Ruiming Ma, ChaoYi Chen, Enhong |
| author_facet | Zhang, Jiaqing Wang, Hao Yin, Mingjia Chen, Bo Jia, Qinglin Zhou, Rui Tang, Ruiming Ma, ChaoYi Chen, Enhong |
| contents | Modern deep recommender models are trained under a continual learning paradigm, relying on massive and continuously growing streaming behavioral logs. In large-scale platforms, retraining models on full historical data for architecture comparison or iteration is prohibitively expensive, severely slowing down model development. This challenge calls for data-efficient approaches that can faithfully approximate full-data training behavior without repeatedly processing the entire evolving data stream. We formulate this problem as \emph{streaming dataset distillation for recommender systems} and propose \textbf{DIET}, a unified framework that maintains a compact distilled dataset which evolves alongside streaming data while preserving training-critical signals. Unlike existing dataset distillation methods that construct a static distilled set, DIET models distilled data as an evolving training memory and updates it in a stage-wise manner to remain aligned with long-term training dynamics. DIET enables effective continual distillation through principled initialization from influential samples and selective updates guided by influence-aware memory addressing within a bi-level optimization framework. Experiments on large-scale recommendation benchmarks demonstrate that DIET compresses training data to as little as \textbf{1-2\%} of the original size while preserving performance trends consistent with full-data training, reducing model iteration cost by up to \textbf{60$\times$}. Moreover, the distilled datasets produced by DIET generalize well across different model architectures, highlighting streaming dataset distillation as a scalable and reusable data foundation for recommender system development. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_24958 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DIET: Learning to Distill Dataset Continually for Recommender Systems Zhang, Jiaqing Wang, Hao Yin, Mingjia Chen, Bo Jia, Qinglin Zhou, Rui Tang, Ruiming Ma, ChaoYi Chen, Enhong Information Retrieval Modern deep recommender models are trained under a continual learning paradigm, relying on massive and continuously growing streaming behavioral logs. In large-scale platforms, retraining models on full historical data for architecture comparison or iteration is prohibitively expensive, severely slowing down model development. This challenge calls for data-efficient approaches that can faithfully approximate full-data training behavior without repeatedly processing the entire evolving data stream. We formulate this problem as \emph{streaming dataset distillation for recommender systems} and propose \textbf{DIET}, a unified framework that maintains a compact distilled dataset which evolves alongside streaming data while preserving training-critical signals. Unlike existing dataset distillation methods that construct a static distilled set, DIET models distilled data as an evolving training memory and updates it in a stage-wise manner to remain aligned with long-term training dynamics. DIET enables effective continual distillation through principled initialization from influential samples and selective updates guided by influence-aware memory addressing within a bi-level optimization framework. Experiments on large-scale recommendation benchmarks demonstrate that DIET compresses training data to as little as \textbf{1-2\%} of the original size while preserving performance trends consistent with full-data training, reducing model iteration cost by up to \textbf{60$\times$}. Moreover, the distilled datasets produced by DIET generalize well across different model architectures, highlighting streaming dataset distillation as a scalable and reusable data foundation for recommender system development. |
| title | DIET: Learning to Distill Dataset Continually for Recommender Systems |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2603.24958 |